Chapter 6 Cloud Computing Energy Saving Technique Exercises
6.1 Multiple choice questions
1. The energy consumption composition of the cloud data center includes (D).
①Energy consumption of server clusters ②Energy consumption of network facilities ③Energy consumption of storage devices ④Energy consumption of power supply equipment
A. ①②③ B. ①②④ C. ②③④ D. ①②③④
2. The main evaluation index points of the cloud data center do not include (C).
A. Reliability B. Energy efficiency C. Scale D. Carbon emissions
3. (A) is an indicator generally accepted and adopted by cloud data centers at home and abroad to measure the energy efficiency of cloud data center infrastructure. Its value is the total power consumption of cloud data centers divided by the power consumption of IT equipment.
A. PUE B. DCEP C. CUE D. WUE
4. For information systems such as cloud computing, (D) in the following options is not related to energy-saving optimization technologies.
A. Shutdown technology B. Sleep technology C. Dynamic voltage frequency adjustment technology D. Virtual device technology
5. Steps in the orderly data aggregation process do not include (B).
A. Data migration B. Data compression C. Node deployment D. Data backup
6. Among the following options, (D) is not a typical de-duplication technology.
A. File segmentation B. Fingerprint value calculation C. Data storage D. Data cleaning
7. Among the following options, (C) is not the main strategy for deduplication.
A. Chunking method B. Chunking granularity C. Data diversity D. Metadata processing
8. Among the following options, (C) is mainly responsible for the work of synchronizing the image file and operation log of the backup metadata.
A. Client B. Metadata server C. Secondary metadata server D. Storage node
6.2 Fill in the Blank
1. The actual meaning of PUE is to calculate how much power is actually used (IT equipment) in the total power provided to the cloud data center. The value range of PUE value is generally ( [1.0, ∞) ).
2. The cloud data center must be equipped with environmental control facilities that can adjust (temperature) and (humidity) to ensure the normal operation of the cloud data center.
3. Data de-duplication technology compares the unique characteristics of data such as fingerprint values, and only keeps one copy of the same data. The purpose is to eliminate (data redundancy) and reduce storage capacity requirements
6.3 Short answer questions
1. What are the current main evaluation indicators for cloud data centers?
answer:
The main evaluation indicators of cloud data centers include: reliability, energy efficiency, carbon emissions, water resources, land resources, pollution emissions, and resource recycling.
2. What aspects can be started to realize the green cloud data center?
answer:
- Infrastructure: Try to continuously introduce new technologies for energy saving and environmental protection, and adopt energy-efficient infrastructure to support the deployment of green cloud data centers.
- IT equipment: Reducing the energy consumption of computing equipment in the calculation process can improve the energy utilization rate of IT equipment from the source.
- Energy utilization: Utilizing convergence technology and virtualization technology to improve the energy utilization of green cloud data centers can effectively improve the overall energy efficiency of cloud data centers.
- Energy consumption management: Real-time and comprehensive monitoring of the energy consumption of the entire cloud data center and even the network, multi-dimensional analysis of the massive energy consumption data generated every day, and reasonable energy-saving suggestions are given, and targeted energy efficiency optimization strategies are designed .
3. For cloud computing and other information systems, what are the energy-saving optimization technologies mainly used at present?
answer:
The energy-saving optimization technologies mainly adopted include low-power hardware, shutdown/sleep technology, dynamic voltage frequency adjustment technology, green network communication, temperature control energy-saving technology, virtualization technology, resource allocation, energy-saving scheduling technology, and green data deployment mechanism.
6.4 Answer questions
1. Please explain and analyze the technical connotation of green computing.
answer:
Green computing complies with the needs of building a low-carbon society and is an important aspect of promoting sustainable social development and technological progress. Based on the principle of being responsible for the environment, using computers and related resources, Green Computing emphasizes reducing resource consumption and properly disposing of electronic waste. Green computing involves system structure, system software, parallel distributed computing, and computer networks. It is based on the premise of ensuring the efficiency and reliability of the computing system and providing universal services, and aims at the low energy consumption of the computing system. It emphasizes the use of efficient and energy-saving CPUs, Servers and peripheral devices are oriented to new computer architectures and new computing models including cloud computing. By building energy-aware computing systems, network interconnection environments, and computing service systems, they provide personalized and diverse information that is increasingly pervasive. Services provide a low-power supporting environment.
2. Please explain the principle and existing problems of using virtualization technology to realize cloud computing energy saving.
answer:
Virtualization technology is an important way to realize cloud computing energy saving. By abstracting physical resources into virtual resources, virtualization technology can virtualize multiple virtual machines on a physical host, assign several tasks to run on these virtual machines, and improve the utilization rate of host resources to reduce The number of hosts required reduces energy consumption. In addition, virtual machine migration technology can be used to aggregate virtual machines, thus providing support for shutdown/hibernation technology.
Virtualization itself has to pay a high cost of energy efficiency, and the deeper the level of virtualization, the higher the cost of energy consumption. Therefore, only using the existing virtualization technology, the optimization effect in terms of cloud computing system performance and energy efficiency is limited. Existing virtual machine managers cannot communicate energy consumption characteristics with multiple operating systems supported by the upper layer, nor can they perceive the load and operating status of upper-layer applications, resulting in an unsatisfactory energy efficiency ratio during task scheduling.